[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Draft Minutes 21st August 2013
Minutes 21st August 2013
Attendees
US Department of Defense (DoD) Michael Behrens Voting Member
Oracle Mark Carlson Voting Member
Oracle Martin Chapman Chair
Fujitsu Limited Jacques Durand Voting Member
Cloudsoft Corporation Limited Alex Heneveld Voting Member
Oracle Anish Karmarkar Voting Member
Oracle Ashok Malhotra Voting Member
Rackspace Hosting, Inc. Adrian Otto Secretary
Vnomic Derek Palma Voting Member
Oracle Gilbert Pilz Voting Member
Red Hat Krishna Raman Voting Member
Fujitsu Limited Tom Rutt Voting Member
Software AG, Inc. Prasad Yendluri Voting Member
Intro:
Alex assumes Scribe duties.
Roll: Attendees listed above, 13 of 13 (100%), meeting is quorate.
Agenda:
V2 agenda adopted as posted
Minutes:
14th August 2013: https://www.oasis-open.org/apps/org/workgroup/camp/email/archives/201308/msg00038.html
MOTION: gil: motion to approve minutes from 14 aug as posted, s: ashok. approved w/o objections.
Administrivia:
Cloud plug fest updates?
none
Martin taking LOA at start of sept. will need to discuss chair at next meeting.
#topic Timeline
pretty good
Public Review: now live
gil: make noise about the spec, so that it gets used
mark: tie it to other hype -- eg paas
gil: post blogs. and let him know so the tweeters can tweet.
mark: retweeting is good to get followers for @camp_oasis
Editing team update:
gil: wd 21 is out
gil: trying new way of using kavi
martin: skeptical re use of revisions
gil: advantage is it let's you see the latest easily
martin: but make sure to include WD# clearly
adrian: also likes it. why martin is this a bad idea?
Issue discussion:
https://tools.oasis-open.org/issues/browse/CAMP-71
MOTION gil: motion to resolve 71 with the proposal as posted with a minor (non-material) language change. s: alex
delete "is a" in last sentence
approved w/o obj
https://tools.oasis-open.org/issues/browse/CAMP-31 - jacques
Jacques (Fujitsu): proposal: define the standard interface and placeholder for "states", but let Providers - or applications - define these.
Tom Rutt (Fujitsu): perhaps if a vendor defines an operation, it can also define some extension attributes for the state values for that component
martin: interesting discussion; jacques: will think about moving towards a proposal
https://tools.oasis-open.org/issues/browse/CAMP-80 - gil
good discussion: no action at present but a better understanding all round
https://tools.oasis-open.org/issues/browse/CAMP-29
Mark Carlson (DMTF): Platform components *may* have other *hidden* dependencies - being managed by the platform administrator - not exposed to the user.
Michael Behrens: Consumer view consistency is good
Ashok Malhotra (Oracle): The arguments have all been made ... ownership, management and access control are different
@Ashok if i say i need a new database for my app, i believe i own it and can start/stop it. however the platform creates it. so is it an App Component or a Platform Component ?
Mark Carlson (DMTF): I do think we could document the differences, contexts, etc. between the two - better in the spec.
@Mark +1
derek: not convinced there are 2 types. it might be 4 or 5 or 6 ...
tom: ownership can be amended. but other things, like managementUri
@tom - mgmtUri is already optional; some components have them, and some do not. i do not think there is an issue adding it. some "app components" might have a mgmtUri.
prasad: @Alex, it (db) is still a platform component ,App gets to use (and manage) an instance of it?
Mark Carlson (DMTF): This is the way you could pop out of CAMP and manage how the Platform Service is realised - i.e. through CAMP.
Mark Carlson (DMTF): CIMI
Tom Rutt (Fujitsu): a component or template can be owned by either a platform or an assembly. That would be known by having the reverse pointer to owner which is in the application component attributes. The differences are the application components have their dependencies on other application and platform components, but the platform components currently have no dependency links. Also the platform component has a management uri
@prasad - however if it is dedicated, it should be tied to the App. if i look at the Assembly i should see the PC for the DB. but currently i don't, i only see AC's.
Anish Karmarkar: @tom I believe we should allow PCs (if they stay separate) to depend on other PCs
Gil Pilz (Oracle): +1 Anish - especially as it may involve billing
@Mark - would be great if you can take a stab at clarifying it. maybe the distinctions will then become clear.
(or maybe you'll come to the (my) conclusion that there are a lot of different dimensions here and it isn't a simply binary split)
+1 Anish PCs depending on PCs. also Assemblies linking to PCs. and possibly PCs linking back to Assemblies if it is created for that Assembly.
Anish Karmarkar: one Q I have on billing (and this was brought up by Alex): it is possible to have two ACs point to the same PC. In that case, AC-specific billing wrt usage of that PC would have to be associated with the ACs not the common PC
@Anish - good point. let's say i'm hosting 2 WAR's in a single appserver. i'd expect their API usage to be metered, and possibly billed against that.
Anish Karmarkar: @alex right
https://tools.oasis-open.org/issues/browse/CAMP-83 - jacques
jacques: quite a few new normative assertions needed
Put on agenda for next week.
AOB:
Stragglers: add Prasad
Meeting Adjourned.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]